summary

Article

Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9- Enhanced Gene Targeting

Graphical Abstract

Highlights d iPSC modeling of SCID

d CRISPR/Cas-enhanced gene correction

d Broad T cell receptor repertoire

d No off-target modifications

Authors

Chia-Wei Chang, Yi-Shin Lai,

Erik Westin, ..., Lawrence S. Lamb, Jr.,

Frederick D. Goldman, Tim M. Townes

Correspondence [email protected]

In Brief Using SCID patient-specific induced

pluripotent stem cells (iPSCs) and a T cell

in vitro differentiation system, Chang

et al. define a block in early T cell

development of JAK3-deficient cells.

Correction of the JAK3 mutation by

CRISPR/Cas9-enhanced gene targeting

restores normal T cell development.

Chang et al., 2015, Cell Reports 12, 1–10 September 8, 2015 ª2015 The Authors http://dx.doi.org/10.1016/j.celrep.2015.08.013

Cell Reports

Article

Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting Chia-Wei Chang,1,5 Yi-Shin Lai,1,5 Erik Westin,1,5 Alireza Khodadadi-Jamayran,1,5 Kevin M. Pawlik,1,5

Lawrence S. Lamb, Jr.,2,3,5 Frederick D. Goldman,4,5 and Tim M. Townes1,5,* 1Department of Biochemistry and Molecular Genetics 2Department of Medicine, Division of Hematology/Oncology 3Cell Therapy Lab 4Department of Pediatrics, Division of Hematology/Oncology 5UAB Stem Cell Institute Schools of Medicine and Dentistry, University of Alabama at Birmingham, Birmingham, AL 35294, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.celrep.2015.08.013 This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

SUMMARY

Mutations of the Janus family kinase JAK3 gene cause severe combined immunodeficiency (SCID). JAK3 deficiency in humans is characterized by the absence of circulating T cells and natural killer (NK) cells with normal numbers of poorly functioning B cells (T–B+NK–). Using SCID patient-specific induced pluripotent stem cells (iPSCs) and a T cell in vitro dif- ferentiation system, we demonstrate a complete block in early T cell development of JAK3-deficient cells. Correction of the JAK3 mutation by CRISPR/ Cas9-enhanced gene targeting restores normal T cell development, including the production of mature T cell populations with a broad T cell receptor (TCR) repertoire. Whole-genome sequencing of corrected cells demonstrates no CRISPR/Cas9 off- target modifications. These studies describe an approach for the study of human lymphopoiesis and provide a foundation for gene correction therapy in humans with immunodeficiencies.

INTRODUCTION

Severe combined immunodeficiency (SCID) describes patients with severe defects in T cells with or without accompanying defects in B cells. Naturally occurring mutations in the JAK3 gene (autosomal recessive) and the X-linked common gamma chain gC gene (IL-2RG) are the most common T

–B+ immunodefi- ciencies. Gene ablation experiments in mice demonstrate that Jak3 is critical for early T cell differentiation and Jak3 knockout mice were found to have severely reduced numbers of mature B cells in the bone marrow and in the periphery (Nosaka et al., 1995; Thomis et al., 1995). The similarity of phenotypes of gc and JAK3 SCID suggests that the primary function of JAK3 is to transduce signals from gc-dependent cytokine receptors (IL-2R,

IL-4R, IL-7R, IL-9R, IL-15RA, and IL-21R) to transcription factors (STATs) that activate downstream genes. Mice lacking Jak3 have a profound decrease in thymus cellularity. However, the residual thymocytes proceed to develop into mature T cells and reconsti- tutetheperipheral population.Incontrast,JAK3-deficientpatients have few, if any, peripheral T cells (Notarangelo et al., 2001; Rus- sell et al., 1995).To date, little is known about how JAK3mutations affect human lymphocyte progenitor development. Allogeneic hematopoietic stem cell transplantation is currently

the only established therapy for SCID; however, delayed immune recovery and graft-versus-host disease present significant risks (Pai et al., 2014). Treatment by retroviral-based gene therapy has been successfully demonstrated for X-linked SCID (Hacein-Bey- Abina et al., 2014) and ADA-SCID (Ferrua et al., 2010). However, severe adverse effects of insertional mutagenesis have been observed with retroviral gene therapy (Hacein-Bey-Abina et al., 2003, 2008). Self-inactivating lentiviral vectors have been used effectively in recent clinical trials, but long-term follow-up is needed to thoroughly address safety concerns (Aiuti et al., 2009; Biffi et al., 2013; Sauer et al., 2014). An alternative thera- peutic strategy is one in which patient-specific induced plurip- otent stem cells (iPSCs) are derived, and disease-causing mutations are corrected by gene targeting (Takahashi et al., 2007; Takahashi and Yamanaka, 2006; Yu et al., 2007). These corrected iPSCs could then be differentiated into hematopoietic progenitors for transplantation into patients to treat the disease (Hanna et al., 2007). The recent development of CRISPR/Cas9- enhanced gene targeting dramatically advances the practicality of this strategy (Cong et al., 2013; Mali et al., 2013). IPSC technology combined with in vitro differentiation sys-

tems also provides a powerful platform to recapitulate in vivo development. We and other groups have shown that OP9 stromal cells transduced with Notch ligand Delta-like-1 or Delta-like-4 (OP9-DL1/4) can contribute to a T cell inductive environment (Chang et al., 2014; Schmitt et al., 2004; Schmitt and Zúñiga-Pflucker, 2002). OP9-DL1/4 cells efficiently induce T lymphopoiesis from iPSC-derived CD34+ hematopoietic pro- genitor cells (HPCs), and these HPCs can be differentiated into

Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors 1

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

functional CD8 T cells (Dervovic et al., 2012). Despite the lack of proper major histocompatibility complex (MHC) expression on OP9 cells, the OP9-DL1/4 system has provided a valuable method to study early T cell commitment and thymocyte matu- ration in vitro (de Pooter and Zúñiga-Pflucker, 2007).

In this paper, we demonstrate that differentiation of JAK3-defi- cient human T cells is blocked at an early developmental stage. Similar to previous studies in mouse models, JAK3-deficient early human T cell progenitors undergo apoptosis at a high rate due to low expression of BCL2. We also demonstrate that correc- tion of the human JAK3 mutation by CRISPR/Cas9-enhanced gene targeting restores the differentiation potential of early T cell progenitors. These corrected progenitors are capable of producing NK cells and mature T cell populations expressing a broad repertoire of T cell antigen receptors (TCRs). These studies establish a powerful system for determining the mechanism of immunodeficiency in human SCID patients and for testing phar- macological and genetic therapies for the disorder.

RESULTS

JAK3-Deficient Human T Cells Express Low Levels of BCL2 and Die at an Early Developmental Stage IPSCs were generated from skin keratinocytes (Chang et al., 2009) of a SCID patient homozygous for a C > T nucleotide substi-

tution in exon 14 of the JAK3 gene. This mutation replaces a CGA codon (arginine at 613) with a TGA stop codon (p.R613X). The 4-month-old patient presented with a T–B+NK– clinical phenotype (see Experimental Procedures). To determine whether this SCID phenotype can be recapitulated in vitro, we attempted to differen- tiate patient-specific iPSCs to T lymphocytes using our previously published two-step OP9 and OP9-DL4 system (Chang et al., 2014). JAK3-deficient iPSCs grew at a rate comparable to control iPSCs derived from healthy donors, and these iPSCs efficiently differentiated into CD34+ hematopoietic progenitors (HPs) on OP9 stromal cell monolayers. However, when the JAK3-deficient, iPSC-derived CD34+ HPs were plated onto OP9-DL4 stromal monolayers, NK and T cell differentiation was dramatically decreased compared to controls (Figure 1A). Only a small popu- lation of CD7+CD16–CD56– T cells or CD7+CD16+CD56+ NK cells was observed at T cell induction day 14 (Figure 1A). The transitions of early T cell progenitors (ETPs) from

CD4–CD8– double-negative (DN) / CD4+CD8+ double-positive (DP) / CD4+ single-positive (SP) or CD8+ single-positive (SP) T cells are directed by precise activation and repression of spe- cific transcription factors (Rothenberg et al., 2008) (Figure 1B). In control cells (Figure 1C), silencing of PU.1 gene expression and induction of GATA3 and BCL11B gene expression direct early hematopoietic progenitors to proceed to the onset of T lineage commitment. In JAK3-deficient cells, expression of the PU.1

ETP DN DP

Control JAK3

C1837T

CD7

CD 16

/5 6

0.11

72.3

4.56 0.38 0.03

3.46

A B

Control JAK3

C1837T

CD3

TC R α

β

1.70 0

D

CD8

CD 4

Control JAK3

C1837T 6.97 16.3

43.332.7

0 0

1.7398.3

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

PU.1 GATA3 BCL11B RAG1 RAG2 PTCRA

Control JAK3 C1837T

ra #o

to G

AP D

H

C

CD4 SP

CD8 SP

CD7 CD4 CD8

CD3 αβTCR

NK

CD56 CD16

*

*

*

* *

*

Figure 1. In Vitro Differentiation of JAK3 C1837T Patient iPSCs Recapitulates SCID Phenotypes (A) Flow cytometry of iPSC-derived T cells. JAK3

WT iPSCs (Control) and JAK3-deficient iPSCs

(JAK3 C1837T) were differentiated into CD34+ cells

on OP9 stromal cells and, subsequently, into NK

and T cells on OP9-DL4 monolayers. T and NK cell

differentiation from JAK3-deficient iPSCs was

dramatically decreased compared to controls; only

a small population of CD7+CD16–CD56– T cells or

CD7+CD16+CD56+ NK cells was observed at T cell

induction day 14.

(B) Diagram of early T cell development. In the

thymus, early T cell progenitors (ETP) differen-

tiate from CD4–CD8– double-negative (DN) to

CD4+CD8+ double-positive (DP) and, subse-

quently, to CD4+ or CD8+ single-positive stages. In

an alternative pathway, early T cell progenitors

differentiate into NK cells; the dotted line in-

dicates multiple steps in the pathway to mature

CD7+CD16+CD56+ NK cells.

(C) RT-qPCR assays for transcripts of key genes

that regulate early events during specification of

the T cell lineage. RNA levels are shown relative

to GAPDH expression. Data are shown as the

mean ± SD. *p < 0.05 (t test).

(D) Flow cytometry of iPSC-derived T cells. JAK3

WT iPSCs (Control) and JAK3-deficient iPSCs

(JAK3 C1837T) were differentiated into CD34+ cells

on OP9 stromal cells and, subsequently, into T cells

on OP9-DL4 monolayers. At T cell induction day 28,

no CD4+CD8+ DP, CD4+ SP, or CD8+ SP T cells

were detected (top), and neither CD3 or TCRab

were expressed in JAK3-deficient iPSC-derived

T cells (JAK3 C1837T) (lower).

2 Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

gene was not completely silenced, induction of the GATA3 and BCL11B genes was significantly lower than controls, and T cell specification was severely limited. However, low-level expres- sion of the T cell-specific genes RAG1, RAG2, and PTCRA sug- gests that JAK3-deficient cells can progress at low efficiency to an early T cell progenitor stage. Jak3 knockout (KO) mice have a small thymus due to a block in thymocyte differentiation at the CD4–CD8– double-negative 2 (DN2) stage prior to productive TCR rearrangement. Interestingly, some residual thymocytes in Jak3 knockout (KO) mice develop into mature T cells and recon- stitute the peripheral population (Eynon et al., 1999); this does not occur in human JAK3-deficient patients. To further under- stand the developmental defects resulting from JAK3 deficiency in humans, we assayed T cell lineage commitment and matura- tion in JAK3-deficient cells compared to normal JAK3 WT controls. IPSC-derived CD34+ cells were plated onto OP9-DL4 monolayers, and cells were harvested and analyzed for lympho-

A B

0

2

4

6

8

10

12

14

16

TD10 TD17

Ap op

to si

s %

Cntl JAK3

0

0.002

0.004

0.006

0.008

BAX

1 12 2 Control JAK3

C1837T

0

0.005

0.01

0.015

0.02

0.025

0.03

BCL2

ND 1 12 2

Control JAK3 C1837T

Control JAK3 C1837T

C

GFP+ (BCL2 +)GFP- (BCL2 low)

CD3

CD 16

/5 6

CD8

CD 4

0

0 0

2.0

0 5.1

ra #o

to G

AP D

H

BCL2

GAPDH G

FP -

G FP

+

* * Figure 2. BCL2 Partially Rescues T Cell Developmental Defects in JAK3-Deficient, In-Vitro-Derived Cells (A) Apoptosis of JAK3-deficient, iPSC-derived

T cells compared to JAK3 WT controls. Annexin

V-positive cells were analyzed at T cell induction

day 10 (TD10) and 17 (TD17). Four independent

experiments were performed with control JAK3 WT

cells (Control, light blue), and five independent

experiments were performed with JAK3-deficient

cells (JAK3 C1837T, dark blue). Data are shown as

the mean ± SD. *p < 0.005 (t test).

(B) RT-qPCR assays for anti-apoptotic BCL2 and

pro-apoptotic BAX expression in two lines (1 and 2)

from JAK3 WT (Control) and JAK3-deficient cells

(JAK3 C1837T). ND, not detected. RNA levels are

shown relative to GAPDH expression.

(C) Flow cytometry of JAK3-deficient iPSC-derived

T cells transduced with BCL2-2A-GFP lentivirus to

assess effects on NK (CD16+56+) and T cell (CD3+)

development and on CD4+CD8+ DP to CD4+ SP or

CD8+ SP T cell maturation. (top right of C) Western

blot of BCL2 protein in GFP– and GFP+ sorted

populations.

cyte markers at T cell induction day 28. No CD4+CD8+ DP, CD4+ SP, or CD8+ SP T cells were detected in JAK3-deficient cells. Moreover, CD3 and TCR ab were not significantly detected in JAK3-defi- cient cells (Figure 1D). The complete absence of CD4+CD8+ cells indicates that human JAK3-deficient cells arrest at the CD4–CD8– DN stage. Low expression of the DN3-associated genes PTCRA, RAG1, and RAG2 suggests that human JAK3-deficient cells arrest at or before the DN2 stage. Unlike the results in Jak3 KO mice in which residual thymocytes develop past the CD4–CD8– DN stage into mature T cells that partially reconsti- tute the peripheral blood system, our

data demonstrate a complete block of T cell development in hu- man JAK3-deficient cells. This block is consistent with the absence of mature T cells in peripheral blood of JAK3-deficient human SCID patients. The profound defects in lymphocyte development of JAK3-

deficient cells can be explained by the absence of IL-7 signaling, which plays an important role in lymphoid progenitor survival (Li et al., 2004; von Freeden-Jeffry et al., 1997) and differentiation (Kang et al., 1999). IL-7/JAK3 signaling maintains thymocyte ho- meostasis by regulating the BCL2 family of apoptotic regulators. Thymocytes and peripheral T cells from Jak3 KO mice have a high apoptotic index in part through selectively elevating Bax, a pro-apoptotic factor, and by reducing expression of Bcl2, an anti-apoptotic factor (Wen et al., 2001). Similarly, we observed an increase in apoptosis of in-vitro-derived human JAK3-defi- cient cells compared to controls at T cell induction day 10 (9%–2.2%) and T cell induction day 17 (7%–1.9%) (Figure 2A).

Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors 3

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

Consistent with this phenotype, BAX levels were increased and BCL2 levels were reduced in JAK3-deficient cells compared to controls (Figure 2B).

Forced expression of Bcl2 rescues T but not B or NK cell development in gc-deficient mice (Kondo et al., 1997). Trans- plantation of Jak3 KO mice with Bcl2-expressing Jak3 KO bone marrow cells also improves peripheral T cell numbers (Wen et al., 2001). To determine whether overexpression of BCL2 rescues the T cell developmental defects of human JAK3-deficient cells, we transduced in-vitro-derived, JAK3-defi- cient CD34+ cells with a lentivirus containing a BCL2-2A-GFP polycistron driven by the EF1a promoter. After transduction, CD34+ cells were plated onto OP9-DL4 monolayers and assayed for NK and T cell markers at T cell induction day 28. No CD3–CD16+CD56+ NK cells were found in GFP– (JAK3–; BCL2 low) or GFP+ cells (JAK3–; BCL2+) (Figure 2C). These findings are consistent with reports demonstrating that the absence of NK cells in gc-deficient and Jak3-deficient mice is due to the lack of functional IL-15 signaling (Giri et al., 1994) and is indepen- dent of Bcl2-mediated anti-apoptosis. CD3+ cells were only detected in GFP+ (JAK3–; BCL2+) cells suggesting that BCL2 released the developmental block at the DN stage in JAK3-defi- cient cells. Interestingly, a second developmental arrest was evident at the DP stage; no further differentiation of CD8+CD4+

DP cells was observed in GFP+ cells (Figure 2C). These data are consistent with recent demonstrations that signaling by intra- thymic IL-7 is necessary for CD8 lineage specification of DP thy- mocytes (Hare et al., 2000; Park et al., 2010).

In summary, the studies described above demonstrate that human SCID phenotypes can be recapitulated in vitro with patient-derived iPSCs. JAK3 deficiency results in proliferative defects in DN thymocytes. Forced expression of BCL2 enhances survival of DN cells, which further differentiate into DP thymocytes. Nevertheless, DP thymocytes fail to mature to SP T cells, and this defect may result from the absence of IL-7/ JAK3 signaling.

Correction of the JAK3 Deficiency in SCID Human Induced Pluripotent Stem Cells by CRISPR/ Cas9-Enhanced Gene Targeting To determine whether normal T cell development can be restored in JAK3-deficient SCID patient cells, we corrected the JAK3 mutation in iPSCs by CRISPR/Cas9-enhanced gene tar- geting. Six guide RNAs within introns upstream and downstream of exon 14 were designed to target wild-type Cas9 or the D10A Cas9 nickase near the C1837T mutation, and a correction tem- plate was used for homology-directed repair (HDR) (Figure 3A). IPSCs were nucleofected with two plasmids expressing the D10A Cas9 nickase and paired guide RNAs or a single plasmid expressing wild-type Cas9 and a single guide RNA. Cells were grown in medium containing G418 for 2 weeks post nucleofec- tion. Individual colonies were picked, expanded, and genotyped by PCR (Figure 3B, top). The efficiency of CRISPR/Cas9-medi- ated JAK3 gene correction is shown in Figure 3C. Three clones from wild-type Cas9 + gRNA #1, three clones from wild-type Cas9 + gRNA #2, and six clones from D10A Cas9 nickase + paired gRNAs #1 and #2 were further verified by Sanger sequencing. In 12 sequenced clones, two homozygous cor-

rected clones (one clone from D10A Cas9 nickase + paired gRNA #1 and #2 and one clone from wild-type Cas9 + gRNA #1) and ten heterozygous corrected clones were identified (Fig- ure 3D). Restoration of JAK3 gene expression was demonstrated by RT-PCR (JAK3 mRNA) (Figure 3B, lower-left panel) and west- ern blot (JAK3 protein) (Figure 3B, lower right).

Specificity of CRISPR/Cas9-Directed JAK3 Correction The potential for off-target, CRISPR/Cas9-directed genome modifications raises some concerns about the use of this approach for therapy in humans. In cancer cell lines, relatively high levels of off-target mutagenesis by Cas9-gRNAs have been described (Fu et al., 2013). However, several groups have recently demonstrated by whole-genome sequencing (WGS) that off-target modifications are rare in human iPSCs and human embryonic stem cells (Smith et al., 2014; Veres et al., 2014). To determine the specificity of CRISPR/Cas9-directed JAK3 correction in human SCID iPSCs, we performed WGS before and after gene correction. The genomes of two heterozygous and one homozygous corrected clones were sequenced. The two heterozygous clones were corrected with gRNA #2 + wild- type Cas9, and the homozygous clone was corrected with gRNA #1 + gRNA #2 + D10A nickase Cas9. The 20-base CRISPR guide sequences were mapped to the

human reference genome, allowing up to three mismatches in order to identify potential off-target sites (Table 1; Figure S1). These sites were then analyzed for variations in the iPSC samples following CRISPR/Cas9-directed gene correction. Whole-genome sequencing of the one homozygous and two heterozygous corrected iPSC lines demonstrated that no muta- tions (SNVs nor indels) were introduced into the 1,450 potential off-target sites (Table 2). These results demonstrate the speci- ficity of CRISPR/Cas9-directed gene correction.

Restoration of T Cell Development after CRISPR/ Cas9-Directed JAK3 Correction To determine whether T cell development is restored after JAK3 gene correction, NK and T cell generation were verified, and T cell lineage commitment and maturation were analyzed. We and other groups have previously demonstrated in the OP9-DL4 in vitro system that T cell differentiation sequentially passes through intermediates observed in vivo: CD34+CD7+

T/NK committed stage; CD7+CD4+CD8! immature, SP stage; CD4+CD8+ DP stage; and finally, CD3+CD8+ TCRab mature stage. Mature T cells are polyclonal, proliferate, and secrete cy- tokines in response to mitogens (Chang et al., 2014; Timmer- mans et al., 2009). Therefore, control, JAK3-deficient, and JAK3-corrected human iPSCs were differentiated into hemato- poietic progenitors on OP9 monolayers, and CD34+ cells were positively selected with anti-CD34 magnetic beads. These cells were plated onto OP9-DL4 monolayers, and non-adherent cells were analyzed for lymphocyte markers at T cell induction day (TD) 14, 21, 28, and 35 (Figure 4). In normal controls (green line), 1.2 3 107 CD7+ cells (84% of cells counted in the lymphoid gate) were generated at T cell induction day 14 from 1–2 3 106

CD34+ cells. At this stage, about 20% of CD7+ cells were CD7+CD16+CD56+ NK cells (2.4 3 106). T cell markers CD4, CD8, CD3, and TCR ab were sequentially detected upon T cell

4 Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

maturation. As we reported previously, at T cell induction day 35 the NK population decreased (6.9 3 104), and more than 50% of residual cells were CD8 SP cells (1.2 3 106). In JAK3-deficient cells (blue line), only 4.5 3 104 CD7+ cells (38.9% of cells counted in lymphoid gate) were generated at T cell induction day 14 from 1–2 3 106 CD34+ cells. The number of total CD7+ cells decreased during extended culture and T cell markers CD3, CD4, CD8, and TCR ab were not significantly expressed. Similar to control cells, 1–2 3 106 CD34+ JAK3-corrected cells

differentiated into 4.7 3 106 CD7+ cells (91% of cells counted in lymphoid gate) at T cell induction day 14 including 4.3 3 105

CD7+CD16+CD56+ NK cells. After further differentiation to TD21, TD28, and TD35, T cell maturation markers CD3, CD4, CD8, and TCR ab were abundantly observed (Figure 4). Our data indicate that JAK3-corrected iPSCs generate NK and T cells with similar efficiency. To determine whether TCR rearrangement is reestablished in

JAK3-corrected T cells, TCR Vb typing was performed by flow

18 19

JAK3 genomic locus

* 1410 11 12 13 15 16 17 20 21

PGKNeo

Colonies examined

PCR posi#ve colonies

%

gRNA #1 39 9 23

gRNA #2 45 33 73.3

gRNA #3 16 1 6.25

gRNA #4 9 3 33.3

gRNA #5 3 0 0

gRNA #6 7 0 0

gRNA #1 + #2 14 14 100

gRNA #3 + #4 3 0 0

gRNA #5 + #6 4 0 0

A

B

D

C

5’ PCR

3’ PCR

JAK3 pa#ent C1837T

Heterozygous corrected

Homozygous corrected

1 2 3 4 5 6

JAK3

GAPDH

0.15 0 0.08 JAK3/ GAPDH

Figure 3. CRISPR/Cas9-Enhanced Correc- tion of the JAK3 C1837T Mutation in Pa- tient-Specific iPSCs (A) Strategy for genome modification using

CRISPR/Cas9 to induce double-strand breaks in

the JAK3 locus and a correction template for

homology-directed repair. Top line, structure of the

JAK3 gene. Open boxes, exons. Red asterisk,

C1837T mutation. Blue arrows, guide RNAs.

(B) (top) PCR analysis demonstrating homologous

recombination; primers for 50 and 30 analysis are

indicated by green arrows. (lower left) RT-PCR

analysis demonstrating JAK3 mRNA expression in

JAK3 WT (Control), JAK3-deficient (JAK3 C1837T),

and corrected (JAK3 Corrected) T cells. (lower

right) Western blot analysis demonstrating JAK3

protein expression in JAK3 WT (Control), JAK3-

deficient (JAK3 C1837T), and corrected (JAK3

Corrected) T cells. JAK3/GAPDH ratios were

quantified with ImageLab software (Bio-Rad).

(C) Summary of targeting efficiencies of guide

RNAs.

(D) Sanger sequencing of the PCR amplicons from

parental JAK3 iPSCs (left), heterozygous corrected

(middle), and homozygous corrected iPSCs (right).

The heterozygous clone was corrected with wild-

type Cas9 + gRNA#2, and the homozygous clone

was corrected with D10A Cas9 nickase + gRNA#1

and #2.

cytometry and summarized in Figure 5A. JAK3-corrected T cells expressed all 21 of the Vb segments that we tested there- fore, a broad TCR repertoire was restored. Finally, we examined the integrity of the TCR signaling pathway, a surrogate of T cell function, in JAK3-corrected T cells by measuring cell surface activation markers following anti-CD3/CD28 stimu- lation. On day 3 post-stimulation, the percentage of CD3+CD25+CD69+ T cells increased from 0.68% to 59.7% in JAK3-

corrected T cells, similar to the increase observed in control cells (0.01%–37.6%) (Figure 5B). These data and results described above demonstrate that correction of the JAK3 C1837T (p.R613X) mutation by CRISPR/Cas9-enhanced gene targeting in an in vitro iPSC model system restores normal T cell develop- ment with the capacity to produce functional, mature T cell pop- ulations with a broad TCR repertoire.

DISCUSSION

Muchhasbeenlearned aboutlymphopoiesisinthepast2decades from naturally occurring immunodeficient mice and from knockout and knockin mouse models. These data have provided funda- mental knowledge about the mechanisms involved in lymphocyte development and activity. In humans, the phenotype of lympho- cytes in the peripheral blood of SCID patients has been well described, but studies on critical steps of lymphoid commitment and thymocyte development have been difficult to perform.

Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors 5

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

Access to bone marrow and thymocyte samples from untreated patients with SCID is challenging since these conditions are rare and infants typically present with life-threatening infections requiring urgent HSC transplantation to survive. The strategy that wedescribefor studyinghuman SCIDbypassesthese restrictions; large numbers of hematopoietic progenitors can be produced from patient-specific iPSCs in vitro, and the mechanisms respon- sible for immunodeficiency can be precisely determined. In this study, we demonstrate that T cell development in human JAK3- deficient SCID is completely blocked before or at the CD4–CD8–

(DN2) stage. Interestingly, forced expression of BCL2 enhances survival of DN cells, which further differentiate into DP thymocytes. However, DP thymocytes fail to mature to SP T cells, and this defect may result from the absence of IL-7/JAK3 signaling.

We also demonstrate that correction of the human JAK3 mutation by CRISPR/Cas9-enhanced gene targeting restores the differentiation potential of early T cell progenitors. Corrected progenitors are capable of producing NK cells and mature T cell populations expressing a broad TCR repertoire. Whole-genome sequencing of one homozygous and two heterozygous cor- rected iPSC lines demonstrates that no mutations (SNVs nor indels) are introduced into 1,450 potential off-target sites, sug- gesting a strong specificity for CRISPR/Cas9-directed gene correction. In summary, these studies describe an approach for the study of human lymphopoiesis and provide a foundation for gene correction therapy in humans with immunodeficiencies

and other monogenic disorders. For gene therapy, we envision transplantation of the CD34+ cells that are generated in the first phase of in vitro culture. These cells include early multipotent he- matopoietic progenitors that generate all myeloid and erythroid cells in colony forming assays in addition to the lymphoid cells that we describe here. Our results suggest that there are no intrinsic defects in lineage specification of early hematopoietic progenitors produced in vitro. However, we have not been able to generate all of these lineages after transplantation into immu- nodeficient (NSG) mice. These results suggest that human he- matopoietic progenitors produced in vitro do not home to mouse bone marrow niches that support self-renewal. However, these early progenitors may be incorporated into human bone marrow niches to which endogenous progenitors naturally home. After safety studies are completed in NSG mice, phase 1 clinical trials will be required to determine whether these early progenitors are capable of engraftment and sustained reconstitution of multiline- age hematopoiesis in human patients.

EXPERIMENTAL PROCEDURES

Patient Information The patient was enrolled in an institutional review board-approved study, and

parents signed consents, in accordance with the Declaration of Helsinki. The

family history was negative for immune deficiencies. For the first 8 months of

age, he had poor weight gain, diarrhea, and recurrent bronchiolitis requiring

frequent hospitalization. He was admitted to the hospital at 8 months of age

with severe respiratory distress and oral thrush. Bronchoscopy with bronchial

alveolar lavage demonstrated bacterial (pseudomonas, H flu, S. pneumonia)

and viral organisms (respiratory syncytial virus). Immunologic evaluations

demonstrated severe hypogammaglobulinemia, with an immunoglobulin E

(IgE) <3 ml/ml, IgA <4 mg/dl, IgG = 29 mg/dl, IgM = 26 mg/dl. Immune pheno-

typing of peripheral blood demonstrated complete absence of CD3+ T cells

and NK cells, though B cells were present (absolute B cell count = 875 ml).

Mitogen studies demonstrated a complete lack of response to concanavalin

A, pokeweed mitogen and phytohemagglutinin A. The diagnosis of SCID

was confirmed by genetic testing, which indicated a homozygous C > T nucle-

otide substitution in exon 14 of the JAK3 gene, resulting in the replacement of

an arginine codon (CGA) with a stop codon (TGA) at amino acid position 613.

This is the first report linking this JAK3 variant (rs149316157) to a clinical case

of SCID. The patient underwent a reduced intensity conditioning matched un-

related bone marrow transplant and is doing well now two years off therapy

with complete immune reconstitution.

Human iPSC Reprogramming and Characterization For iPSC induction, 5 3 104 primary keratinocytes were seeded into one well of

a 6-well plate. On the following day, keratinocytes were transduced with 1 ml of

virus supernatant and 1 ml of human keratinocyte medium containing poly-

brene at a final concentration of 4 mg/ml. The keratinocytes were spinfected

Table 1. Identification of Potential Off-Target Sites

gRNA#l gRNA#2

Sequence GTGAGATACAGATACAGACA AATGATTTGCCTGGAATGCC

Target sites (0 mismatch) 1 1

1 base mismatch, potential off-target sites 3 0

2 base mismatch, potential off-target sites 80 13

3 base mismatch, potential off-target sites 1,109 243

Total 1,193 257

Potential off-target sites were identified by aligning the CRISPR/Cas9 guide sequences to the hg19 reference genome using EMBOSS fuzznuc

software (v.6.6.0.0) (Rice et al., 2000) and allowing for a maximum of three mismatches.

Table 2. Variant Analysis of Potential Off-Target Sites

Clone 1 Clone 2 Clone 3

Cas9 wild-type wild-type D10A nickase

gRNA gRNA#2 gRNA#2 gRNA#l+gRNA#2

JAK3 C1837T genotype C/T C/T C/C

Discordant variants in

off-targets and flanking

100 bases

0 0 0

Whole-genome sequencing (WGS) was performed on three corrected

iPSC clones and the uncorrected control (see Whole-Genome

Sequencing and Analysis in Experimental Procedures). Variants (from

the reference genome) that were common to all four iPSC samples

were excluded from further analysis. The remainder were screened to

determine whether these variants were located in potential off-target

sites. No variants were observed in potential off-target sites. BEDTools

(v.2.17.0) (Quinlan and Hall, 2010) was used to search for non-excluded

variants in potential off-target sites.

6 Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

at 800 3 g for 45 min (day 1). The transduction procedure was repeated again

the next day. On day 3, cells were changed to fresh human keratinocyte me-

dium and cultured for 2 more days. On day 5, the keratinocytes were trypsinized

and transferred to a 10-cm dish pre-seeded with mitomycin C-treated murine

embryonic fibroblasts (MEFs) and cultured in human keratinocyte medium.

On day 7, cells were changed to human embryonic stem (ES) medium and

continuously cultured on the same dish for 3–4 weeks. ES medium was

changed daily. Potential iPSC colonies were visible after 2–3 weeks. These col-

onies were individually picked and expanded on MEFs for analysis. To remove

the integrated lentiviral and polycistronic sequences, iPSCs were infected with

a Cre-expressing adenovirus (rAd-Cre-IE). Individual colonies were picked and

Cre-mediated removal of floxed sequences was verified by PCR using the

primers gctaattcactcccaaagaagacaag and cttcagcaagccgagtcctg (Figure S2).

Generation of CD34+ Cells and T Cells with OP9 Co-culture The procedure was described previously (Chang et al., 2014) with the following

modifications. Cultures of human induced pluripotent stem cells (hiPSCs) in one

well of 6-well plate were treated as described by Ohnuki and Yamanaka (2009)

with collagenase-trypsin-knockout serum replacement (CTK-KSR) solution to

make small cell clumps. Cell clumps were then transferred to a 10-cm plate

that was pre-seeded with 2-day old OP9 cells in a-MEM-based medium con-

taining 10% FBS, 1 3 penicillin/streptomycin and 100 mM mono-thioglycerol.

A

0

2

4

6

8

10

TD14 TD21 TD28 TD35 0

2

4

6

8

TD14 TD21 TD28 TD35

0

2

4

6

8

TD14 TD21 TD28 TD35

CD7+ CD16+CD56+

Lo g1

0 C

el l N

um be

r

Lo g1

0 C

el l N

um be

r

CD3+

0

2

4

6

8

TD14 TD21 TD28 TD35

TCRαβ

Lo g1

0 C

el l N

um be

r

Lo g1

0 C

el l N

um be

r

0

2

4

6

8

TD14 TD21 TD28 TD35

CD4+ CD4+CD8+

0

2

4

6

8

TD14 TD21 TD28 TD35

CD8+

Lo g1

0 C

el l N

um be

r

Lo g1

0 C

el l N

um be

r

Lo g1

0 C

el l N

um be

r

0

2

4

6

8

TD14 TD21 TD28 TD35

JAK3 C1837T

JAK3 Corrected

Control

Figure 4. In Vitro Differentiation of JAK3- Corrected Patient iPSCs Produces NK Cell and T Cells with Phenotypic and Functional Characteristics of Mature T Cells Expression of NK cell and T cell developmental

markers in JAK3 WT (Control, green, n = 3), JAK3-

deficient (JAK3 C1837T, blue, n = 5), and JAK3-

corrected (JAK3 Corrected, orange, n = 6) T cells.

Cells were stained with the indicated antibodies

and analyzed by flow cytometry at T cell induction

days 14, 21, 28, and 35 (TD 14, 21, 28 and 35). Data

are shown as the mean ± SD.

Themediumwaschangedeveryotherday,andcells

were cultured for 18 days without any splitting. After

18 days of co-culture, cells were harvested by treat-

ing with dissociation solution (0.15% collagenase IV

and 0.015% hyaluronidase in a-MEM medium) for

about 30 min and followed by 0.25% trypsin for

another 30 min. CD34+ cells were then purified on

anti-CD34+ magnetic beads (MicroBead Kit; Milte-

nyi Biotec). For T cell differentiation, these CD34+

cells were plated onto OP9-DL4 cells and cultured

with a-MEM medium containing 20% FBS,

5 ng/ml hFlt3-L, 5 ng/ml hIL-7, and 10 ng/ml hSCF.

The medium was changed every other day, and

cells were transferred to new OP9-DL4 plates every

4 days. After completion of our studies, Menon et al.

(2015) described the correction of SCID-X1 patient-

derived iPSCs by TALENs and the generation of

mature NK cells and T cell precursors.

T Cell Stimulation In vitro derived T cells from hiPSCs were stimulated

by incubation with CD3/28 beads (Invitrogen) ac-

cording to the manufacturer’s protocol for 3 days

prior to analysis by flow cytometry, as previously

described (Chang et al., 2014).

Flow Cytometry Cells were harvested and washed before analysis

with an LSRFortessa cell analyzer (BD Bioscience).

For cell surface staining, propidium iodide (PI, Sigma-Aldrich) was used to

exclude dead cells. For the apoptosis assay, harvested cells were first stained

with cell surface antibodies for 30 min. After washing once with 1 3 PBS, the

cells were resuspended in 100 ml of Annexin Binding Buffer (Invitrogen) con-

taining Annexin V-647 (Invitrogen) and PI and incubated for 15 min before add-

ing 400 ml of Annexin Binding Buffer with PI. Antibodies were obtained from BD

Biosciences unless otherwise indicated: CD3 (Percp-Cy5-5, clone UCHT1),

CD4 (PE-Cy7, clone SK3), CD7 (APC, BV510, clone M-T701), CD8 (APC-

Cy7, clone SK1), CD16 (PE, clone B73.1), CD25 (FITC, clone 2A3), CD34

(PE-Cy7, clone WM59), CD43 (PE, clone 1G10), CD56-PE (clone MY31),

CD69 (FITC, clone L78), TCR-ab (FITC, PE, clone T10B9.1A-31), Beta Mark

TCR Repertoire Kit (Beckman Coulter).

Vector Construction The polycistronic OSKM vector was previously described (Chang et al., 2009).

The Lenti-hDL4-mCherry plasmid was constructed by cloning a PCR-ampli-

fied human DL4 cDNA (Open Biosystems), an IRES fragment (Open Bio-

systems) and mCherry cDNA into a lentiviral vector (pDL171) which contains

the EF1a promoter. PCRs were performed using PrimeStar polymerase

(Takara). To construct CRISPR plasmids, we cloned designed gRNA oligos

into pX330 and pX335 plasmids following the Zhang lab protocol (Addgene).

The list of primers for construction of gRNA sequences is as follows: (1)

Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors 7

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

gRNA-F1: caccGTG AGA TAC AGA TAC AGA CA, (2) gRNA-R1: aaacTGT CTG

TAT CTG TAT CTC AC, (3) gRNA-F2: caccgAAT GAT TTG CCT GGA ATG CC,

(4) gRNA-R2: aaacGGC ATT CCA GGC AAA TCA TTc, (5) gRNA-F3: caccg

CAG CCT AGG CAA AGG CCT GC, (6) gRNA-R3: aaacGCA GGC CTT TGC

CTA GGC TGc, (7) gRNA-F4: caccgTGC CAA CAG AAC TGC CTG AT, (8)

gRNA-R4: aaacATC AGG CAG TTC TGT TGG Cac, (9) gRNA-F5: caccGAC

CAG GGT GCA AGT GTG GA, (10) gRNA-R5: aaacTCC ACA CTT GCA CCC

TGG TC, (11) gRNA-F6: caccGCT CCT CAG CCT GGC ATT CA, and (12)

gRNA-R6: aaacTGA ATG CCA GGC TGA GGA GC. To construct the JAK3

repair plasmid, wild-type human genomic DNA was PCR amplified using

JAK3 primer sets (50 arm, forward: gtcgacgtcgacgctcagtgaagctgaagtattcctt

ctgcttcacagggcgaccactac and 50 arm, reverse: atttaaatcctcccctcgaacccttac

caaactcctatgcatactacag; 30 arm, forward: ttaattaattaattagcattttaggttcaggttgt

gagaacactagaagagaacaagtca and 30 arm, reverse: gtatacgtatacgcatacctg

gagaggggacaaggtcttgagatgcgagggt).

After digesting with enzymes (50 arm, SalI and SwaI; 30 arm, PacI and

BstZ17I), the PCR products were cloned into a plasmid containing a LoxP-

PGK-Neo-LoxP fragment. All of the oligos used in this study were synthesized

by Integrated DNA Technologies (IDT). To construct the BCL2 lentiviral

plasmid, a primer set (forward: agccaccttaattaagccaccatggcgcacgctggga

gaacggggtacgata and reverse: taacagagagaagttcgtggctccggatcccttgtggcc

cagataggcacccagggtgat) was used to amplify the human BCL2 cDNA (Open

Biosystems) fragment. The amplified product was linked to GFP through a

2A sequence by PCR and cloned into the pDL171 vector.

Cell Culture IPSCs were cultured on mitomycin-C-treated MEFs derived from E14.5 CF-1

embryos in ES cell medium consisting of DMEM F-12 supplemented with 1 3

non-essential amino acids, 1 3 penicillin-streptomycin, 1 3 L-glutamine (all

from Mediatech), 20% Knockout Serum Replacement (Invitrogen), 2-ME

(Sigma), and 5–10 ng/ml bFGF (Invitrogen). Human primary keratinocytes

were cultured in DermaLife K Medium Complete Kit (LifeLine Cell Technology).

OP9 cells were purchased from ATCC and grown in a-MEM medium with 20%

FBS and penicillin-streptomycin. OP9-DL4 cells were established by trans-

ducing OP9 cells with a lentivirus containing hDL4 and mCherry.

Virus Production For preparation of lentivirus, 10 mg of the lentiviral vector, 2.5 mg of the enve-

lope plasmid (pMDG), and 7.5 mg of the packaging plasmid (pCMBVdR8.9.1)

were co-transfected into 5 3 106 293T cells by Fugene 6 (Roche or Promega).

Virus-containing supernatant was collected 2 days after transfection and

passed through a 0.45-mm filter.

Gene Targeting IPSCs were treated with 0.25% trypsin for 5 min to generate single-cell sus-

pensions. After washing twice with 1 3 PBS, 1–2 million cells were mixed

with 5 mg of JAK3 repair plasmid and 5 mg of either pX330-JAK3 or pX335-

JAK3 plasmid for Nucleofection (Human Stem Cell Nucleofector Kit, program

A-023, Lonza) and subsequent plating onto MEFs. Two to 4 days later, hES

medium containing 30 mg/ml of G418 was added to the plates to select for

drug-resistant colonies. The colonies were picked 3–4 weeks post selection

and expanded for genomic DNA extraction. For genotyping, a 50 primer set

(tgctaaagcgcatgctccagact and gtcttcatctcagggtcggct) and a 30 primer set

(cctctctgtgcattatggcag and gccttctatcgccttcttg) were used. To remove

the Neo selection marker, iPSCs were infected with a Cre-expressing adeno-

virus (rAd-Cre-IE) and later analyzed by PCR for marker loss using the

following primers: Jak3-Creout-F ttgggagtgggctctgtagtatgc and Jak3-Cre-

out-R ttcttcctgcccagcctcgtcatt (Figure S2).

RT-PCR Total RNA was isolated from in-vitro-derived cells with Trizol reagent (Invitro-

gen). cDNA was synthesized from 0.5–2 mg of total RNA using the Superscript

First-strand Synthesis System (Invitrogen) according to the manufacturer’s

instructions. SYBR Green PCR Master Mix (Life Technologies) was used for

qPCR according to the manufacturer’s instructions. All values were normalized

relative to GAPDH expression. Primer sets used for qPCR are GAPDH (F:

actcctccacctttgacgct, R: tcccctcttcaagggtctacatg); PU.1 (F: gtgcaaaatggaag

ggtttc, R: ggagctccgtgaagttgttc); GATA3 (F: tgtttcctttcactggccaca, R: aacggc

aactggtgaacggta); BCL11B (F: ggcgatgccagaatagatgccg, R: ccaggccacttggc

tcctctatctccaga); RAG1 (F: ccttactgttgagactgcaatatcc, R: ctgaagtcccagtatat

acttcacac); RAG2 (F: cccagaagcagtaataatcatcgag, R: atgtgggatgtagtagatc

ttgc); pTa (F: gggtcttacctcagcagttac, R: cctcacacagtgtgacgcag); BCL2 (F:

gactgagtacctgaaccggc, R: gggccaaactgagcagagtc); BAX (F: aagaccagggtg

gttgggac, R: gtaagaaaaatgcccacgtc); and JAK3 (F: agtcagacgtctggagcttc,

R: gtgagcagtgaaggcatgagtc).

Whole-Genome Sequencing and Analysis DNA from iPSCs was sheared using a Covaris S2 Focused-ultrasonicator:

130-ml samples in microTUBEs were subjected to two 60-s cycles of 10%

duty cycle, intensity of 4, and 200 cycles per burst in frequency sweeping

mode. DNA Chip (DNA 1000 Kit; Agilent Technologies) analysis using an Agi-

lent 2100 Bioanalyzer indicated an average fragment size of 400 bp. Library

preparation was performed using an NEBNext Ultra DNA Library Prep Kit for

Illumina (NEB #E7370), and the final library concentration was determined by

qPCR using a KAPA Illumina Library Quantification Kit (KK4835; KAPA Bio-

systems) and an Applied Biosystems ViiA 7 Real-Time PCR System (Life

Technologies).

Sequencing clusters were produced on the flow cell using an Illumina

TruSeq PE Cluster Kit v.3-cBot-HS (PE-401-3001) and an Illumina cBot.

WGS was performed using an Illumina TruSeq SBS Kit v.3-HS-200 cycles

(FC-401-3001) and an Illumina HiSeq 2500 upgrade to generate 2 3 100

single-index paired-end reads for bioinformatic analysis.

s#mulated un-s#mulated

Control

JAK3 Corrected

CD 25

CD69

62.8

32.3

54.1

37.6 0.07

4.08

0.25

23.0

0

1

2

3

4

5

6

7

TCR Vβ familyA

B

% o

f C D

3+ c

el ls

0

0.07

0.01

0.05

s#mulated (isotype control)

Figure 5. In Vitro Differentiation of JAK3-Corrected Patient iPSCs Produces T Cells with Phenotypic and Functional Characteristics of Mature T Cells (A) T cell receptor (TCR) Vb analysis of JAK3-corrected T cells. A highly diverse

repertoire of TCR Vb is represented in T cells derived from corrected SCID

patient iPSCs.

(B) Flow cytometry demonstrating T cell activation in JAK3-corrected

T cells. T cells derived from JAK3 WT (Control) and JAK3-corrected

iPSCs were stimulated with anti-CD3/28 beads for 3 days before analysis

of activation markers CD25 and CD69. Matched isotype antibodies were

used as negative controls (isotype). These data were gated on CD3+

populations.

8 Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

Potential off-target sites were identified by aligning the CRISPR/Cas9 guide

sequences to the hg19 reference genome using EMBOSS fuzznuc software

(v.6.6.0.0) (Rice et al., 2000) and allowing for a maximum of three mismatches;

1,193 sites were identified for the first guide sequence (GTGAGATACAGATA

CAGACA) and 257 sites for the second guide sequence (AATGATTTG

CCTGGAATGCC).

All of the reads from the WGS for each sample were mapped to the hg19

reference genome using the BWA (v.0.7.5a) mem algorithm (Li and Durbin,

2010) and duplicate reads were removed using Picard tools (v.1.100) (http://

broadinstitute.github.io/picard/). Local realignment and base quality re-cali-

bration were performed using GATK (v.2.7-2) (McKenna et al., 2010). Both

SNVs and indels were called using GATK HaplotypeCaller. Additionally,

SNVs and indels were separately re-calibrated as described in GATK Best

Practices, and quality filters were applied. Genome STRiP v.2.0 (Handsaker

et al., 2011, 2015) was used to call the large deletions against a background

population of 100 samples from the 1000 Genomes Project. Inversions, dupli-

cations, and translocations were called using Delly v.0.6.3 (Rausch et al.,

2012), and insertions were called using Pindel v.0.2.5a8 (Ye et al., 2009).

Variants from the reference genome that were common to all four iPSC sam-

ples were excluded from CRISPR/Cas9 off-target analysis. The non-excluded

variants were screened using BEDTools (v.2.17.0) (Quinlan and Hall, 2010) to

determine whether they fell within the potential off-target sites ±100 flanking

base pairs. The analysis demonstrated that none of these variants reside

in the off-target sites and suggests that these variants were randomly

accumulated.

All of the functional variants (excluded and non-excluded) with a low allele

frequency (<1% in dbSNP 138, 1000 Genomes, and NHLBI-ESP 6500),

conserved in phastCons 46-way elements (Siepel et al., 2005) and with a

high CADD score (CADD R10) (Kircher et al., 2014) were then annotated using

the ANNOVAR software package (Wang et al., 2010) and screened for known

associations with diseases in ClinVar (v.20140902) (Landrum et al., 2014),

GWAS Catalog (Welter et al., 2014), and COSMIC (v.70) (Forbes et al.,

2008); additionally, we manually screened the variants with a high CADD score

(CADD >20) in HGMD (Stenson et al., 2003). None of these variants, including

the JAK3 C1837T variant, were found to be associated with disease in

the databases queried. Unlike the other variants, however, the JAK3

C1837T (p.R613X) variant (rs149316157) is predicted to be significantly dele-

terious, with a GERP score (Davydov et al., 2010) of 3.85 and a CADD score

of 36; we report here the association of this JAK3 variant with a clinical case

of SCID.

ACCESSION NUMBERS

The WGS data can be accessed at the NCBI Sequence Read Archive data-

base with the accession number SRP056149.

SUPPLEMENTAL INFORMATION

Supplemental Information includes two figures and can be found with this

article online at http://dx.doi.org/10.1016/j.celrep.2015.08.013.

Received: March 13, 2015

Revised: June 2, 2015

Accepted: August 4, 2015

Published: August 27, 2015

REFERENCES

Aiuti, A., Cattaneo, F., Galimberti, S., Benninghoff, U., Cassani, B., Callegaro,

L., Scaramuzza, S., Andolfi, G., Mirolo, M., Brigida, I., et al. (2009). Gene ther-

apy for immunodeficiency due to adenosine deaminase deficiency. N. Engl. J.

Med. 360, 447–458.

Biffi, A., Montini, E., Lorioli, L., Cesani, M., Fumagalli, F., Plati, T., Baldoli, C.,

Martino, S., Calabria, A., Canale, S., et al. (2013). Lentiviral hematopoietic

stem cell gene therapy benefits metachromatic leukodystrophy. Science

341, 1233158.

Chang, C.W., Lai, Y.S., Pawlik, K.M., Liu, K., Sun, C.W., Li, C., Schoeb, T.R.,

and Townes, T.M. (2009). Polycistronic lentiviral vector for ‘‘hit and run’’

reprogramming of adult skin fibroblasts to induced pluripotent stem cells.

Stem Cells 27, 1042–1049.

Chang, C.W., Lai, Y.S., Lamb, L.S., Jr., and Townes, T.M. (2014). Broad T-cell

receptor repertoire in T-lymphocytes derived from human induced pluripotent

stem cells. PLoS ONE 9, e97335.

Cong, L., Ran, F.A., Cox, D., Lin, S., Barretto, R., Habib, N., Hsu, P.D., Wu, X.,

Jiang, W., Marraffini, L.A., and Zhang, F. (2013). Multiplex genome engineering

using CRISPR/Cas systems. Science 339, 819–823.

Davydov, E.V., Goode, D.L., Sirota, M., Cooper, G.M., Sidow, A., and Batzo-

glou, S. (2010). Identifying a high fraction of the human genome to be under

selective constraint using GERP++. PLoS Comput. Biol. 6, e1001025.

de Pooter, R., and Zúñiga-Pflücker, J.C. (2007). T-cell potential and develop-

ment in vitro: the OP9-DL1 approach. Curr. Opin. Immunol. 19, 163–168.

Dervovic, D.D., Ciofani, M., Kianizad, K., and Zúñiga-Pflücker, J.C. (2012).

Comparative and functional evaluation of in vitro generated to ex vivo CD8

T cells. J. Immunol. 189, 3411–3420.

Eynon, E.E., Livák, F., Kuida, K., Schatz, D.G., and Flavell, R.A. (1999). Distinct

effects of Jak3 signaling on alphabeta and gammadelta thymocyte develop-

ment. J. Immunol. 162, 1448–1459.

Ferrua, F., Brigida, I., and Aiuti, A. (2010). Update on gene therapy for adeno-

sine deaminase-deficient severe combined immunodeficiency. Curr. Opin.

Allergy Clin. Immunol. 10, 551–556.

Forbes, S.A., Bhamra, G., Bamford, S., Dawson, E., Kok, C., Clements, J.,

Menzies, A., Teague, J.W., Futreal, P.A., and Stratton, M.R. (2008). The cata-

logue of somatic mutations in cancer (COSMIC). Curr. Protoc. Hum. Genet.

Chapter 10, Unit 10.11. http://dx.doi.org/10.1002/0471142905.hg1011s57.

Fu, Y., Foden, J.A., Khayter, C., Maeder, M.L., Reyon, D., Joung, J.K., and

Sander, J.D. (2013). High-frequency off-target mutagenesis induced by

CRISPR-Cas nucleases in human cells. Nat. Biotechnol. 31, 822–826.

Giri, J.G., Ahdieh, M., Eisenman, J., Shanebeck, K., Grabstein, K., Kumaki, S.,

Namen, A., Park, L.S., Cosman, D., and Anderson, D. (1994). Utilization of the

beta and gamma chains of the IL-2 receptor by the novel cytokine IL-15.

EMBO J. 13, 2822–2830.

Hacein-Bey-Abina, S., Von Kalle, C., Schmidt, M., McCormack, M.P., Wulf-

fraat, N., Leboulch, P., Lim, A., Osborne, C.S., Pawliuk, R., Morillon, E., et al.

(2003). LMO2-associated clonal T cell proliferation in two patients after gene

therapy for SCID-X1. Science 302, 415–419.

Hacein-Bey-Abina, S., Garrigue, A., Wang, G.P., Soulier, J., Lim, A., Morillon,

E., Clappier, E., Caccavelli, L., Delabesse, E., Beldjord, K., et al. (2008). Inser-

tional oncogenesis in 4 patients after retrovirus-mediated gene therapy of

SCID-X1. J. Clin. Invest. 118, 3132–3142.

Hacein-Bey-Abina, S., Pai, S.Y., Gaspar, H.B., Armant, M., Berry, C.C.,

Blanche, S., Bleesing, J., Blondeau, J., de Boer, H., Buckland, K.F., et al.

(2014). A modified g-retrovirus vector for X-linked severe combined immuno-

deficiency. N. Engl. J. Med. 371, 1407–1417.

Handsaker, R.E., Korn, J.M., Nemesh, J., and McCarroll, S.A. (2011). Discov-

ery and genotyping of genome structural polymorphism by sequencing on a

population scale. Nat. Genet. 43, 269–276.

Handsaker, R.E., Van Doren, V., Berman, J.R., Genovese, G., Kashin, S.,

Boettger, L.M., and McCarroll, S.A. (2015). Large multiallelic copy number

variations in humans. Nat. Genet. 47, 296–303.

Hanna, J., Wernig, M., Markoulaki, S., Sun, C.W., Meissner, A., Cassady, J.P.,

Beard, C., Brambrink, T., Wu, L.C., Townes, T.M., and Jaenisch, R. (2007).

Treatment of sickle cell anemia mouse model with iPS cells generated from

autologous skin. Science 318, 1920–1923.

Hare, K.J., Jenkinson, E.J., and Anderson, G. (2000). An essential role for the

IL-7 receptor during intrathymic expansion of the positively selected neonatal

T cell repertoire. J. Immunol. 165, 2410–2414.

Kang, J., Coles, M., and Raulet, D.H. (1999). Defective development of

gamma/delta T cells in interleukin 7 receptor-deficient mice is due to impaired

expression of T cell receptor gamma genes. J. Exp. Med. 190, 973–982.

Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors 9

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013

Kircher, M., Witten, D.M., Jain, P., O’Roak, B.J., Cooper, G.M., and Shendure,

J. (2014). A general framework for estimating the relative pathogenicity of

human genetic variants. Nat. Genet. 46, 310–315.

Kondo, M., Akashi, K., Domen, J., Sugamura, K., and Weissman, I.L. (1997).

Bcl-2 rescues T lymphopoiesis, but not B or NK cell development, in common

gamma chain-deficient mice. Immunity 7, 155–162.

Landrum, M.J., Lee, J.M., Riley, G.R., Jang, W., Rubinstein, W.S., Church,

D.M., and Maglott, D.R. (2014). ClinVar: public archive of relationships among

sequence variation and human phenotype. Nucleic Acids Res. 42, D980–

D985.

Li, H., and Durbin, R. (2010). Fast and accurate long-read alignment with

Burrows-Wheeler transform. Bioinformatics 26, 589–595.

Li, W.Q., Jiang, Q., Khaled, A.R., Keller, J.R., and Durum, S.K. (2004). Inter-

leukin-7 inactivates the pro-apoptotic protein Bad promoting T cell survival.

J. Biol. Chem. 279, 29160–29166.

Mali, P., Yang, L., Esvelt, K.M., Aach, J., Guell, M., DiCarlo, J.E., Norville, J.E.,

and Church, G.M. (2013). RNA-guided human genome engineering via Cas9.

Science 339, 823–826.

McKenna, A., Hanna, M., Banks, E., Sivachenko, A., Cibulskis, K., Kernytsky,

A., Garimella, K., Altshuler, D., Gabriel, S., Daly, M., and DePristo, M.A. (2010).

The Genome Analysis Toolkit: a MapReduce framework for analyzing next-

generation DNA sequencing data. Genome Res. 20, 1297–1303.

Menon, T., Firth, A.L., Scripture-Adams, D.D., Galic, Z., Qualls, S.J., Gilmore,

W.B., Ke, E., Singer, O., Anderson, L.S., Bornzin, A.R., Alexander, I.E., Zack,

J.A., and Verma, I.M. (2015). Lymphoid regeneration from gene-corrected

SCID-X1 subject-derived iPSCs. Cell Stem Cell 16, 367–372.

Nosaka, T., van Deursen, J.M., Tripp, R.A., Thierfelder, W.E., Witthuhn, B.A.,

McMickle, A.P., Doherty, P.C., Grosveld, G.C., and Ihle, J.N. (1995). Defective

lymphoid development in mice lacking Jak3. Science 270, 800–802.

Notarangelo, L.D., Mella, P., Jones, A., de Saint Basile, G., Savoldi, G.,

Cranston, T., Vihinen, M., and Schumacher, R.F. (2001). Mutations in severe

combined immune deficiency (SCID) due to JAK3 deficiency. Hum. Mutat.

18, 255–263.

Ohnuki, M.T.K., and Yamanaka, S. (2009). Generation and characterization of

human induced pluripotent stem cells. Curr. Protoc. Stem Cell Biol. Chapter 4,

Unit 4A 2.

Pai, S.Y., Logan, B.R., Griffith, L.M., Buckley, R.H., Parrott, R.E., Dvorak, C.C.,

Kapoor, N., Hanson, I.C., Filipovich, A.H., Jyonouchi, S., et al. (2014). Trans-

plantation outcomes for severe combined immunodeficiency, 2000-2009.

N. Engl. J. Med. 371, 434–446.

Park, J.H., Adoro, S., Guinter, T., Erman, B., Alag, A.S., Catalfamo, M., Kimura,

M.Y., Cui, Y., Lucas, P.J., Gress, R.E., et al. (2010). Signaling by intrathymic

cytokines, not T cell antigen receptors, specifies CD8 lineage choice and

promotes the differentiation of cytotoxic-lineage T cells. Nat. Immunol. 11,

257–264.

Quinlan, A.R., and Hall, I.M. (2010). BEDTools: a flexible suite of utilities for

comparing genomic features. Bioinformatics 26, 841–842.

Rausch, T., Zichner, T., Schlattl, A., Stütz, A.M., Benes, V., and Korbel, J.O.

(2012). DELLY: structural variant discovery by integrated paired-end and

split-read analysis. Bioinformatics 28, i333–i339.

Rice, P., Longden, I., and Bleasby, A. (2000). EMBOSS: the European Molec-

ular Biology Open Software Suite. Trends Genet. 16, 276–277.

Rothenberg, E.V., Moore, J.E., and Yui, M.A. (2008). Launching the T-cell-line-

age developmental programme. Nat. Rev. Immunol. 8, 9–21.

Russell, S.M., Tayebi, N., Nakajima, H., Riedy, M.C., Roberts, J.L., Aman, M.J.,

Migone, T.S., Noguchi, M., Markert, M.L., Buckley, R.H., et al. (1995). Mutation

of Jak3 in a patient with SCID: essential role of Jak3 in lymphoid development.

Science 270, 797–800.

Sauer, A.V., Di Lorenzo, B., Carriglio, N., and Aiuti, A. (2014). Progress in gene

therapy for primary immunodeficiencies using lentiviral vectors. Curr. Opin.

Allergy Clin. Immunol. 14, 527–534.

Schmitt, T.M., and Zúñiga-Pflücker, J.C. (2002). Induction of T cell develop-

ment from hematopoietic progenitor cells by delta-like-1 in vitro. Immunity

17, 749–756.

Schmitt, T.M., de Pooter, R.F., Gronski, M.A., Cho, S.K., Ohashi, P.S., and

Zúñiga-Pflücker, J.C. (2004). Induction of T cell development and establish-

ment of T cell competence from embryonic stem cells differentiated in vitro.

Nat. Immunol. 5, 410–417.

Siepel, A., Bejerano, G., Pedersen, J.S., Hinrichs, A.S., Hou, M., Rosenbloom,

K., Clawson, H., Spieth, J., Hillier, L.W., Richards, S., et al. (2005). Evolution-

arily conserved elements in vertebrate, insect, worm, and yeast genomes.

Genome Res. 15, 1034–1050.

Smith, C., Gore, A., Yan, W., Abalde-Atristain, L., Li, Z., He, C., Wang, Y., Brod-

sky, R.A., Zhang, K., Cheng, L., and Ye, Z. (2014). Whole-genome sequencing

analysis reveals high specificity of CRISPR/Cas9 and TALEN-based genome

editing in human iPSCs. Cell Stem Cell 15, 12–13.

Stenson, P.D., Ball, E.V., Mort, M., Phillips, A.D., Shiel, J.A., Thomas, N.S.,

Abeysinghe, S., Krawczak, M., and Cooper, D.N. (2003). Human Gene Muta-

tion Database (HGMD): 2003 update. Hum. Mutat. 21, 577–581.

Takahashi, K., and Yamanaka, S. (2006). Induction of pluripotent stem cells

from mouse embryonic and adult fibroblast cultures by defined factors. Cell

126, 663–676.

Takahashi, K., Tanabe, K., Ohnuki, M., Narita, M., Ichisaka, T., Tomoda, K.,

and Yamanaka, S. (2007). Induction of pluripotent stem cells from adult human

fibroblasts by defined factors. Cell 131, 861–872.

Thomis, D.C., Gurniak, C.B., Tivol, E., Sharpe, A.H., and Berg, L.J. (1995).

Defects in B lymphocyte maturation and T lymphocyte activation in mice lack-

ing Jak3. Science 270, 794–797.

Timmermans, F., Velghe, I., Vanwalleghem, L., De Smedt, M., Van Copper-

nolle, S., Taghon, T., Moore, H.D., Leclercq, G., Langerak, A.W., Kerre, T.,

et al. (2009). Generation of T cells from human embryonic stem cell-derived

hematopoietic zones. J. Immunol. 182, 6879–6888.

Veres, A., Gosis, B.S., Ding, Q., Collins, R., Ragavendran, A., Brand, H., Erdin,

S., Cowan, C.A., Talkowski, M.E., and Musunuru, K. (2014). Low incidence of

off-target mutations in individual CRISPR-Cas9 and TALEN targeted human

stem cell clones detected by whole-genome sequencing. Cell Stem Cell 15,

27–30, Erratum in Cell Stem Cell. 2014 Aug 7;15(2):254. Cowan, Chad A

[added].

von Freeden-Jeffry, U., Solvason, N., Howard, M., and Murray, R. (1997). The

earliest T lineage-committed cells depend on IL-7 for Bcl-2 expression and

normal cell cycle progression. Immunity 7, 147–154.

Wang, K., Li, M., and Hakonarson, H. (2010). ANNOVAR: functional annotation

of genetic variants from high-throughput sequencing data. Nucleic Acids Res.

38, e164. http://dx.doi.org/10.1093/nar/gkq603.

Welter, D., MacArthur, J., Morales, J., Burdett, T., Hall, P., Junkins, H., Klemm,

A., Flicek, P., Manolio, T., Hindorff, L., and Parkinson, H. (2014). The NHGRI

GWAS Catalog, a curated resource of SNP-trait associations. Nucleic Acids

Res. 42, D1001–D1006.

Wen, R., Wang, D., McKay, C., Bunting, K.D., Marine, J.C., Vanin, E.F.,

Zambetti, G.P., Korsmeyer, S.J., Ihle, J.N., and Cleveland, J.L. (2001). Jak3

selectively regulates Bax and Bcl-2 expression to promote T-cell develop-

ment. Mol. Cell. Biol. 21, 678–689.

Ye, K., Schulz, M.H., Long, Q., Apweiler, R., and Ning, Z. (2009). Pindel: a

pattern growth approach to detect break points of large deletions and medium

sized insertions from paired-end short reads. Bioinformatics 25, 2865–2871.

Yu, J., Vodyanik, M.A., Smuga-Otto, K., Antosiewicz-Bourget, J., Frane, J.L.,

Tian, S., Nie, J., Jonsdottir, G.A., Ruotti, V., Stewart, R., et al. (2007). Induced

pluripotent stem cell lines derived from human somatic cells. Science 318,

1917–1920.

10 Cell Reports 12, 1–10, September 8, 2015 ª2015 The Authors

Please cite this article in press as: Chang et al., Modeling Human Severe Combined Immunodeficiency and Correction by CRISPR/Cas9-Enhanced Gene Targeting, Cell Reports (2015), http://dx.doi.org/10.1016/j.celrep.2015.08.013